Skip to main content
PDF4me APIIntegrationsKnowledge BaseExtensionsAPI TesterBlog

AI - Document Parser in Zapier

What this action does

PDF4me AI - Document Parser runs a saved AI Analyzer from the PDF4me dashboard against any document and returns the extracted data as dynamic JSON inside a Zap. One action covers invoices, purchase orders, receipts, contracts, custom forms, lab reports, or anything you describe in a schema. Pick the Analyzer from the AI Analyzer Id dropdown (Parse for a single schema, Classify for many document variants). Fields appear as individually mappable tokens in every downstream step across 6,000+ Zapier apps.

Related Blog Posts(2)
AI-Rename Vendor Invoices in Zapier: A 5-Step Dropbox + PDF4me Workflow
Five-step Zap: Dropbox New File in Folder → Dropbox Find File → PDF4me AI - Document Parser → Code by Zapier Run JavaScript → Dropbox Upload File. Extracts Customer Name with AI and uses it as the output filename. Real screenshots, exact field values, no code beyond a 2-line JSON.parse.
Read post →
Rename Every Vendor Invoice by Supplier Name in Zapier: A 5-Step Dropbox + PDF4me AI Workflow
Five-step Zap for accounts-payable teams. Dropbox watches your invoice intake folder, PDF4me AI - Document Parser pulls the Vendor Name out of every PDF, Code by Zapier turns the parser output into mappable tokens, and Dropbox uploads the file back with the supplier name as the filename. Real screenshots, exact field values, the same sample-vendor-invoice.pdf used in this walkthrough is downloadable.
Read post →

Sample Invoice for Testing​

Use this sample invoice to confirm the action works end-to-end before pointing your Zap at live data. Drop it into Dropbox, Google Drive, or any storage connector, map the file into File, concatenate File Name + File Ext (for example sample_invoice + .pdf), and pick default_invoice_extraction from the AI Analyzer Id dropdown. The action returns a populated Doc Text JSON you can parse downstream.

Authenticating Your API Request​

To access the PDF4me Web API through Zapier, every Zap action needs a valid PDF4me API key on the PDF4me connection. Get or rotate your key from the developer dashboard.

Get your API KeyOpen AI Document Parser Dashboard

Set up your Analyzer first​

Dashboard URLs you will use in this guide:
Create your Analyzer first (required for every platform)
AI - Document Parser runs a saved Analyzer from the PDF4me developer dashboard. Pick Parse for one document schema, or Classify to route many variants (Client ABC, Client XYZ, generic invoice) in a single call. The same Analyzer Id works in Make, Zapier, Power Automate, n8n, and the REST API.

Important Facts You Should Not Miss​

The Analyzer lives in the PDF4me dashboard, the Zap just references it
Define the schema (fields, types, descriptions) once in the Analyzer list and Analyzer detail pages. The Zapier action reads AI Analyzer Id from the dropdown to pick which Analyzer to run. Edit the schema later and every Zap pointing at the same Id picks it up automatically, no Zap rebuild.
File OR File URL, not both
Map File from a previous step (Dropbox / Google Drive / Gmail attachment / Webhook) when the document already lives in a connected app. Use File URL when the document is hosted at a public HTTPS link. Setting both is unnecessary, the action prefers File when present.
Parse vs Classify is decided by the Analyzer, not the action
Same action, two analyzer modes. A Parse Analyzer returns the fields from its single schema. A Classify Analyzer routes the document to the best-matching Schema and returns its Classification Name plus the matched Schema's extracted fields. Switch behaviour by switching the Id.
Use AI Analyzer Id, not AI Analyzer Name (Deprecated)
Zapier shows AI Analyzer Id as a dropdown of Analyzers from your PDF4me account. The older AI Analyzer Name text field is marked (Deprecated). Leave it empty and pick the Analyzer from AI Analyzer Id.
Zapier PDF4me AI - Document Parser Configure tab: File mapped from Dropbox step 2, File URL empty, File Name mapped as sample_invoice plus .pdf, AI Analyzer Id dropdown set to default_invoice_extraction, and AI Analyzer Name Deprecated left empty

Configure tab: map File from a prior step, leave File URL empty, build File Name from File Name + File Ext tokens, pick AI Analyzer Id, and leave AI Analyzer Name (Deprecated) empty.

Parameters​

Dashboard URLs you will use in this guide:

Required in the action: File OR File URL (one of the two), File Name (with extension), and AI Analyzer Id. Leave AI Analyzer Name (Deprecated) empty.

FieldRequiredWhat it doesExample
FileConditionalBinary file from a previous Zap step. Map a file token from Dropbox, Google Drive, OneDrive, Gmail attachment, Slack file, Webhooks by Zapier, or any connector that exposes a file output. Required when File URL is not provided. In the screenshot, step 2 File is mapped from Dropbox.2. File
File URLConditionalPublicly reachable HTTPS URL to the document. Required when File is not provided. Use when the document lives on a public CDN, S3 bucket, or content host. Leave empty when File is mapped.https://example.com/document.pdf
File NameYesSource document filename with extension. Best practice is to concatenate the previous step File Name and File Ext tokens (for example sample_invoice + .pdf becomes sample_invoice.pdf).sample_invoice.pdf
AI Analyzer IdYesDropdown of Analyzers saved in your PDF4me account. Create Analyzers at https://dev.pdf4me.com/dashboard/#/ai-document-parser/ (+ Add), then edit schemas on https://dev.pdf4me.com/ai-document-parser/?id=<analyzer-guid>. Sent to the API as customisationNote. Pick Parse for one schema or Classify for many variants.default_invoice_extraction
AI Analyzer Name (Deprecated)NoLegacy free-text field. Zapier marks this as deprecated. Leave empty and use AI Analyzer Id instead.—

AI Analyzer Id: how Analyzer Ids map to behaviour​

Parse Analyzer IdOne schema, extracts the fields you defined
Best when every document you process shares the same layout family. Output JSON contains exactly the properties you described in fields on the Analyzer, including table arrays for line items.
Classify Analyzer IdMany schemas, routes and extracts in one call
Best when one Zap must handle several variants (Client ABC, Client XYZ, generic invoice). The AI matches the incoming document to the most fitting Schema by its Classification Prompt, then returns the matched Classification Name plus that Schema's extracted fields.
default_invoice_extractionBuilt-in invoice template
A ready-to-use Analyzer for typical vendor invoices. Pick this when you need invoice extraction immediately without building a custom schema. Move to a custom Parse Analyzer when you need specific fields, vendor logic, or table columns the default does not cover.

Output​

The action returns dynamic JSON. Every property defined in your Analyzer schema appears in Zapier's Test result and as an individually mappable token in every downstream step. The exact shape depends on your Analyzer:

  • Top-level keys match the fieldName values in your schema (string, number, date, or table arrays for line items).
  • Classify Analyzers also return the matched Classification Name so you can branch downstream on which Schema matched.
  • No binary file is returned, the AI parser is a JSON-only action.

A few common token names you will likely see, depending on the schema:

TokenTypeTypical content
invoiceNumberStringInvoice / bill / receipt number
invoiceDateDateNormalised date (when fieldMethod: "generate" is set)
vendorNameStringSupplier or vendor name from the header
totalAmountNumberGrand total in the document currency
lineItemsArrayRepeated row data (one object per line, fields you described in the nested fields array)

Quick Setup​

  1. Build your Analyzer first. Open the Analyzer list, click + Add, type a clear Analyzer Id, pick Parse or Classify, save the row, then add a Document Schema on the Analyzer detail page. See the Parse setup guide or Classify setup guide for full schema rules (including fieldType: "table" for nested rows and fieldMethod: "generate" for derived values).
  2. In your Zap click +, search PDF4me, pick AI - Document Parser.
  3. Connect or reuse your PDF4me account.
  4. Map File from a previous step (Dropbox / Google Drive / Gmail attachment / Webhook), or paste a public HTTPS link into File URL.
  5. Map File Name to the previous step's filename tokens (concatenate File Name + File Ext to keep the original name with the right extension).
  6. From AI Analyzer Id pick your Analyzer (for example default_invoice_extraction, client_invoices, or a custom Parse or Classify Analyzer). Leave AI Analyzer Name (Deprecated) empty.
  7. Click Test action. Inspect the returned JSON in the Zapier Test panel.
  8. Map the returned fields into the next action (Google Sheets Create Row, Airtable Create Record, Gmail Send Email, Slack Send Channel Message, or any of 6,000+ Zapier apps).

Workflow Examples​

Workflow ExamplesOpen by default. Four common AI - Document Parser patterns inside Zaps.
Vendor invoices to Google Sheets
  1. Trigger: Gmail New Attachment on the AP inbox.
  2. AI - Document Parser with File mapped from the attachment, File Name from Gmail tokens, AI Analyzer Id default_invoice_extraction (or your custom Parse Analyzer).
  3. Google Sheets Create Row: map InvoiceNumber, VendorName, TotalAmount, InvoiceDate into the AP tracker.
  4. Optional: Slack Send Channel Message to #finance with the values for visibility.

Impact: Hands-off AP intake from any vendor.

Multi-vendor inbox with Classify
  1. Trigger: Dropbox New File in Folder on a mixed-vendor folder.
  2. AI - Document Parser with AI Analyzer Id pointing at a Classify Analyzer (client_invoices) that holds one Schema per vendor.
  3. Zapier Paths branches on the returned Classification Name and sends each result to the matching Airtable base.

Impact: One Zap handles every client without separate flows.

Hosted document by URL
  1. Trigger: Webhooks by Zapier Catch Hook receives a document link from your portal.
  2. AI - Document Parser with File URL set to the link, AI Analyzer Id set to your custom form Analyzer.
  3. Airtable Create Record with mapped fields builds a CRM contact.

Impact: Portal-uploaded forms become CRM data without manual entry.

Smart file rename + archive
  1. Trigger: Dropbox New File in inbox.
  2. AI - Document Parser returns vendorName and invoiceNumber from the schema.
  3. Dropbox Move File to archive/{{vendorName}} and rename to {{vendorName}}_{{invoiceNumber}}.pdf.

Impact: Clean, searchable archive with zero manual file work.

Frequently Asked Questions​

What is the difference between AI - Document Parser and AI-Invoice Parser?+
AI-Invoice Parser uses a fixed schema tuned for invoices (vendor, line items, totals, dates). AI - Document Parser is schema-driven: it runs whatever Analyzer you pick from AI Analyzer Id, built once in the PDF4me dashboard. The same action handles purchase orders, receipts, contracts, custom forms, lab reports, ID cards, or anything else you describe in a schema. Pick AI - Document Parser whenever you need flexibility or a non-invoice document type.
Where do I build the Analyzer?+
At https://dev.pdf4me.com/dashboard/#/ai-document-parser/ (Analyzer list, + Add) and https://dev.pdf4me.com/ai-document-parser/?id=<analyzer-guid> (Analyzer detail for schemas). Two analyzer types: Parse (one schema) and Classify (many variants). Pick the Analyzer from the AI Analyzer Id dropdown in Zapier.
What happened to AI Analyzer Name?+
AI Analyzer Name is deprecated in Zapier. Use the AI Analyzer Id dropdown instead. Leave the deprecated text field empty.
Should I provide File or File URL?+
Provide one or the other. Map File from a previous Dropbox / Google Drive / Gmail / Webhook step when the document lives inside another connected app. Use File URL when the document is hosted at a public HTTPS link. Setting both is unnecessary; the action prefers File when present.
What format is the output and how do I map fields downstream?+
The response is dynamic JSON with one property per field defined in your Analyzer schema (for example customerName, invoiceNumber, totalAmount, lineItems). Zapier exposes those properties in the Test result of this step. Pick the matching property from the dynamic dropdown in any downstream action.
Which file types does the action accept?+
PDF, PNG, JPG, and JPEG. The AI engine handles OCR internally, so scanned PDFs and image formats work the same as native PDFs.
What does default_invoice_extraction give me?+
A built-in Analyzer for typical vendor invoices. Pick it when you need invoice extraction immediately without building a custom schema. Move to your own Parse Analyzer when you need specific fields, vendor logic, or table columns the default does not cover.
Can one Zap process several document variants in one run?+
Yes, with a Classify Analyzer. Build one Classify Analyzer in the dashboard with one Schema per variant (Client ABC, Client XYZ, generic invoice). The AI picks the best-matching Schema at runtime and the action returns its fields plus the Classification Name. Use a Zapier Paths step to branch on the Classification Name and send each variant to the right downstream app.
Will the same Analyzer work from Make, Power Automate, n8n, or the REST API?+
Yes. The Analyzer Id is shared across every platform. Save it once in the dashboard and call it from Zapier, Make, Power Automate, n8n, or the REST API by passing the same Id. Schema changes propagate to every platform without rewiring.

Same Task on Other Platforms​

Get Help​

PDF4me on ZapierGet your API KeyZapier Documentation